Identification of the susceptibility genes is critical to elucidating the pathogenesis of insulin-dependent diabetes mellitus (IDDM) and may provide new approaches to the development of novel intervention and prevention strategies. To date, genome-wide linkage analyses have suggested 14 promising IDDM susceptibility intervals. Whereas localization of these intervals is a major step forward, a difficult task remains in identifying the exact disease genes involved. The investigators have recently observed significant associations between IDDM and two polymorphic markers within the CTLA4 gene on chromosome 2q33 (p less than 10^-5) in a large multi-ethnic collection of diabetic families. These findings provide strong evidence for a least one IDDM susceptibility gene in the region surrounding CTLA4 (designate IDDM12). Analyses of additional markers flanking CTLA4 have narrowed IDDM12 to a region of less than 750kb. Within this interval, CTLA4, and its neighboring locus CS28, are the most likely candidate genes for IDDM12, not only because of their genomic locations but also their important roles in the regulation of T-cell activation. In this application, the investigators propose to continue their ongoing studies to identify IDDM12 using a combination of genetic and functional studies. Specific Aim 1 seeks to further define the genomic interval that contains IDDM12. The investigators propose to identify new polymorphic markers within the 750kb interval from YAC and BAC clones, and to test for their associations with IDDM in their diabetic pedigrees and case/controls. These results will be used to establish a so-called association curve (a plot of strength of association against marker distance), which predicts the likely location of the disease gene. Specific Aim 2 seeks to investigate the potential involvement of CTLA4 and CD28 in IDDM pathogenesis through three sets of experiments. The first two seek the genetic basis for the involvement of the two molecules in IDDM through identification of mutations in the coding and regulatory regions of CTLA4 and CD28. The third will explore the functional importance of the CTLA4/CD28 molecules in IDDM pathogenesis by comparing the expression levels of CTLA4, CD28, cytokines and other immunological parameters in normal controls, prediabetic and diabetic patients stratified by IDDM12 genotypes. Finally, if neither CTLA4 nor CD28 is IDDM12, the investigators will carry out positional cloning studies to identify the disease gene.